Prasad Cheema @_prasad_cheema_

|| Humble ML researcher || Joined October 2015

Tweets

716
Followers

41
Following

222
Likes

2K

Dr Kareem Carr, Statistics Person @kareem_carr

2 years ago

WHY do we divide by n-1 when computing the sample variance? I've never seen this way of explaining this concept anywhere else. Read on if you want a completely new way of looking at this.

168 1K 8K 2.1M 8K

Download Image

Gabriel Peyré @gabrielpeyre

3 years ago

Brunn-Minkowski inequality is one of the fundamental inequalities in convex geometry, which generalizes the isoperimetric inequality. en.wikipedia.org/wiki/Brunn%E2%… terrytao.wordpress.com/tag/brunn-mink…

3 158 885 0 161

Download Video

Clément Canonne (on Blue🦋Sky) @ccanonne_

3 years ago

Let's start with a PSA: print this. Hang it on the wall near your desk. Make it a t-shirt and wear it to work. lkozma.net/inequalities_c… It is the single most useful list of inequalities I have ever encountered, and it's a goddamn treasure. (Even if you only ever use a subset!) 2/

1 9 60 0 29

Download Image

Steven Strogatz @stevenstrogatz

4 years ago

Along with being a great mathematician, Gian-Carlo Rota was a charming gentleman (he was very kind to me when I was a young professor at @MIT). Plus he was a wise, witty, and opinionated writer. Here he shares "Ten Lessons I wish I had been Taught": alumni.media.mit.edu/~cahn/life/gia…

12 116 491 0 164

Download Image

Phan K.S. @MTLevG

5 years ago

This may just be the most chaotic 1st round I've seen in the last year! Singdam Kafeafocus (blue) came back to TKO Singpayak less than 2 minutes after he was knocked down himself.

3 12 48 0 3

Download Video

Frank Nielsen @FrnkNlsn

4 years ago

"The Many Faces of Information Geometry": Gentle short introduction to information geometry. Open access PDF by @amermathsoc "Notices of the AMS". IG aims at unravelling geometric structures of families of probability distributions ams.org/journals/notic… #InformationGeometry

3 104 424 0 154

Download Image

Elvis Dohmatob @dohmatobelvis

4 years ago

Did you know that stochastic gradient-flow on a least-squares objective is equivalent to regularized ridge regression (with parameter lambda ~ 1 / t), along the entire optimization path ?! arxiv.org/abs/2003.07802. Nice work by Alnur Ali, @EdgarDobriban, and Ryan Tibshirani.

6 7 69 0 31

Frank Nielsen @FrnkNlsn

4 years ago

Asymptotic error exponent of probability of error in Bayesian multiple hypothesis testing related to smallest Chernoff information between distinct pairs of hypothesis. Calculated for exponential family hypothesis from Bregman Voronoi diagrams. tinyurl.com/ClosestPairChe…

0 2 26 0 6

Download Image

Gintare Karolina Dziugaite @gkdziugaite

4 years ago

Our NeurIPS 2021 spotlight presents more evidence that CMI is a universal framework for generalization. Joint work led by @HaghifamMahdi, in collaboration with @roydanroy and Shay Moran.

Mahdi Haghifam @HaghifamMahdi

4 years ago

Our NeurIPS 2021 spotlight presents more evidence that CMI is a universal framework for generalization. Joint work led by @HaghifamMahdi, in collaboration with @roydanroy and Shay Moran.

3 39 218 0 70

Download Image

0 27 158 0 31

Timothy Nguyen @IAmTimNguyen

4 years ago

Published on the arxiv today is my response to @PiaMalaney + @EricRWeinstein's work on "Economics as Gauge Theory" which Eric Weinstein presented at UChicago last month: arxiv.org/abs/2112.03460 Summary of the objections is as follows: (1/n)

17 59 473 0 95

Download Image

Frank Nielsen @FrnkNlsn

4 years ago

Save the date: Dec. 14 2021, 10:25 to 11:10 JST Talk: "Information geometry: An invitation for machine learning" neuripsmeetup.jp/2021/ #NeurIPS2021 meetup. Registration is free🙏 Retweet

0 15 49 0 8

Download Image

Gabriel Peyré @gabrielpeyre

4 years ago

Lloyd’s algorithm is the continuous counterpart of k-means. Optimizes the optimal quantization energy. en.wikipedia.org/wiki/Lloyd%27s… en.wikipedia.org/wiki/Vector_qu…

7 84 577 0 103

Download Video

Analysis Fact @AnalysisFact

4 years ago

Unified theory of Fourier transforms johndcook.com/blog/2017/08/2…

0 5 29 0 5

Emre Kıcıman @emrek

4 years ago

The videos of the #causalml sessions at the recent Microsoft Research Summit are now available on-demand microsoft.com/en-us/research… @MSFTResearch #causality #ml

Emre Kıcıman @emrek

4 years ago

The videos of the #causalml sessions at the recent Microsoft Research Summit are now available on-demand microsoft.com/en-us/research… @MSFTResearch #causality #ml

1 17 79 0 11

3 20 130 0 61

Chris McComb (he/him) @ccmccomb

4 years ago

This semester I've been teaching a new class on designing AI/ML products in @CMU_Mech, and as part of that class I have learned SO MUCH about great tools for ML rapid prototyping. Here are a few of my favorites:

10 71 342 0 211

Daniël Lakens @lakens

4 years ago

New blog post: Why p-values should be interpreted as p-values and not as measures of evidence. I had to write response to the recent paper by @StefanieMuff, @eb_nilsen, @BobOHara, and @chloe_nater daniellakens.blogspot.com/2021/11/why-p-… I hope this will spark an interesting discussion!